머신 러닝을 이용한 경제분석 = A Short Guide to Machine Learning for Economists

다국어 초록 (Multilingual Abstract)

With the development of various AI (Artificial Intelligence) techniques and increased availability of big data, ML (Machine Learning) is expected to become the essential technology that would affect many aspects of our economy and society. With this in mind, the purpose of this paper is to provide an overview of ML techniques with emphasis on its application to economics.
Contrasting the key differences in ML techniques and econometric methodologies, we first explain the key techniques used in supervised learning, which are widely used in industry and academia. Then we provide a survey of recent economic research that uses ML techniques and introduce debates on its impact on labor market and the value of data. We conclude with discussing the current limitations of ML technique in terms of economic research, while we believe that ML will fruitfully complement the current methodologies of economics.

국문 초록 (Abstract)

본 논문은 경제학 전공자를 대상으로 인공 지능을 구현하는 핵심 기법인머신 러닝의 개념과 주요 방법론, 경제학과 경제에 미치는 영향을 개괄적으로 소개하고자 한다. 먼저 머신 러닝의 주요 범주인 지도 학습, 비지도학습, 강화 학습의 개념을 소개하고 기존 계량경제학 접근법과의 차이점을설명한다. 그리고 학계와 산업계에서 널리 연구되고 활용되는 지도 학습분야에서 분류 및 회귀를 위해 사용되는 주요 방법론을 예를 통해 설명한뒤 머신 러닝 기법이 활용된 경제학 분야의 최신 연구들, 노동시장에 미치는 영향, 데이터의 가치를 둘러싼 논쟁에 대해 살펴본다.

참고문헌 (Reference)

1 Posner, Eric C., "래디컬 마켓: 공정한 사회를위한 근본적 개혁 (Radical Markets: Uprooting Capitalism andDemocracy for a Just Society)" 부키 2019

2 김수현, "경제분석을 위한 텍스트 마이닝" 26 (26): 2019

3 Matthieu Picault, "Words are not all created equal: A new measure of ECB communication" Elsevier BV 79 : 136-156, 2017

4 Kang, Jun Seok, "Where Not to Eat? Improving Public Policy by Predicting Hygiene Inspections Using Online Reviews" Association for Computational Linguistics 1443-1448, 2013

5 Gentzkow, Matthew, "What Drives Media Slant? Evidence From U.S. Daily Newspapers" The Econometric Society 78 (78): 35-71, 2010

6 Alexander Peysakhovich, "Using methods from machine learning to evaluate behavioral models of choice under risk and ambiguity" Elsevier BV 133 : 373-384, 2017

7 Levitt, Steven D., "Using Electoral Cycles in Police Hiring to Estimate the Effect of Police on Crime" 87 (87): 270-290, 1997

8 Benjamin Moritz, "Tree-Based Conditional Portfolio Sorts: The Relation between Past and Future Stock Returns" Elsevier BV 2016

9 Acemoglu, Daron, "Too Much Data: Prices and Inefficiencies in Data Markets" National Bureau of Economic Research 2019

10 David B. Lobell, "The use of satellite data for crop yield gap analysis" Elsevier BV 143 : 56-64, 2013

11 Surowiecki, James, "The Wisdom of Crowds" Time Warner Books Uk 2005

12 Dave Donaldson, "The View from Above: Applications of Satellite Data in Economics" American Economic Association 30 (30): 171-198, 2016

13 Kleinberg, Jon, "The Theory Is Predictive, but Is It Complete? An Application to Human Perception of Randomness"

14 Athey, Susan, "The Impact of Machine Learning on Economics" University of Chicago Press 507-547, 2019

15 Pearl, Judea, "The Book of Why: The New Science of Cause and Effect" Basic Books 2018

16 Matthew Gentzkow, "Text as Data" American Economic Association 57 (57): 535-574, 2019

17 Kim, Soohyon, "Text Mining for Macroeconomic Analysis (in Korean)" 2020

18 Bholat, David, "Text Mining for Central Banks"

19 Abelson, Brian, "Targeting Direct Cash Transfers to the Extremely Poor" 1563-1572, 2014

20 Vapnik, Vladimir N., "Statistical Learning Theory" Wiley-Interscience 1998

21 Samuel, A. L., "Some Studies in Machine Learning Using the Game of Checkers" 3 (3): 210-229, 1959

22 Arrieta-Ibarra, Imanol, "Should We Treat Data as Labor? Moving beyond ‘Free’" 108 : 38-42, 2018

23 Daron Acemoglu, "Robots and Jobs: Evidence from US Labor Markets" University of Chicago Press 2019

24 Arthur E. Hoerl, "Ridge Regression: Biased Estimation for Nonorthogonal Problems" Informa UK Limited 12 (12): 55-67, 1970

25 McBride, Linden, "Retooling Poverty Targeting Using Out-of-Sample Validation and Machine Learning" 2016

26 Eva Ascarza, "Retention Futility: Targeting High-Risk Customers Might be Ineffective" SAGE Publications 55 (55): 80-98, 2018

27 Hui Zou, "Regularization and variable selection via the elastic net" Wiley 67 (67): 301-320, 2005

28 Robert Tibshirani, "Regression Shrinkage and Selection Via the Lasso" Wiley 58 (58): 267-288, 1996

29 Eric A. Posner, "Radical Markets: Uprooting Capitalism and Democracy for a Just Society" Princeton University Press 2018

30 Aaron Chalfin, "Productivity and Selection of Human Capital with Machine Learning" American Economic Association 106 (106): 124-127, 2016

31 Jon Kleinberg, "Prediction Policy Problems" American Economic Association 105 (105): 491-495, 2015

32 Agrawal, Ajay, "Prediction Machines: The Simple Economics of Artificial Intelligence" Harvard Business Review Press 2018

33 J. Blumenstock, "Predicting poverty and wealth from mobile phone metadata" American Association for the Advancement of Science (AAAS) 350 (350): 1073-1076, 2015

34 Kreif, Noémi, "Oxford Research Encyclopedia of Economics and Finance, by Noémi Kreif and Karla DiazOrdaz" Oxford University Press 2019

35 Bernheim, B. Douglas, "Non-Choice Evaluations Predict Behavioral Responses to Changes in Economic Conditions" National Bureau of Economic Research 2013

36 Dong-Jin Pyo, "News media sentiment and asset prices in Korea: text-mining approach" Informa UK Limited 1-23, 2019

37 Kelly, Bryan, "Measuring Technological Innovation over the Long Run" NBER 2018

38 Lee, Young Joon, "Measuring Monetary Policy Surprises Using Text Mining: The Case of Korea"

39 Baker, Scott R., "Measuring Economic Policy Uncertainty" 131 (131): 1593-1636, 2016

40 J. Vernon Henderson, "Measuring Economic Growth from Outer Space" American Economic Association 102 (102): 994-1028, 2012

41 Lucca, David O., "Measuring Central Bank Communication: An Automated Approach with Application to FOMC Statements" NBER 2011

42 Sendhil Mullainathan, "Machine Learning: An Applied Econometric Approach" American Economic Association 31 (31): 87-106, 2017

43 Susan Athey, "Machine Learning Methods That Economists Should Know About" Annual Reviews 11 (11): 685-725, 2019

44 Mitchell, Thomas M., "Machine Learning" McGraw-Hill, Inc 1997

45 Feigenbaum, James J, "Intergenerational Mobility during the Great Depression"

46 Efron, Bradley, "Institute of Mathematical Statistics Monographs, (Book 5)" Cambridge University Press 2016

47 Acosta, Miguel, "Hanging on Every Word : Semantic Analysis of the FOMC’s Postmeeting Statement" Board of Governors of the Federal Reserve System (U.S.) 2015

48 Géron, Aurélien, "Hands-On Machine Learning with Scikit- Learn and TensorFlow: Concepts, Tools, and Techniques to Build Intelligent Systems" O’Reilly Media 2017

49 Goodfellow, Ian, "Generative Adversarial Nets" Neural Information Processing Systems 2014

50 Susan Athey, "Generalized random forests" Institute of Mathematical Statistics 47 (47): 1148-1178, 2019

51 J. E. Blumenstock, "Fighting poverty with data" American Association for the Advancement of Science (AAAS) 353 (353): 753-754, 2016

52 Acosta, Miguel, "FOMC Responses to Calls for Transparency" Board of Governors of the Federal Reserve System (U.S.) 2015

53 Stefan Wager, "Estimation and Inference of Heterogeneous Treatment Effects using Random Forests" Informa UK Limited 113 (113): 1228-1242, 2017

54 Athey, Susan, "Ensemble Methods for Causal Effects in Panel Data Settings" 109 : 65-70, 2019

55 Gu, Shihao, "Empirical Asset Pricing via Machine Learning" NBER 2018

56 Lee, Young Joon, "Deciphering Monetary Policy Board Minutes Through Text Mining Approach: The Case of Korea" Bank of Korea WP

57 Edward L. Glaeser, "Crowdsourcing City Government: Using Tournaments to Improve Inspection Accuracy" American Economic Association 106 (106): 114-118, 2016

58 N. Jean, "Combining satellite imagery and machine learning to predict poverty" American Association for the Advancement of Science (AAAS) 353 (353): 790-794, 2016

59 Breiman, Leo, "Classification and Regression Trees" Chapman and Hall/CRC 1984

60 Hal R. Varian, "Big Data: New Tricks for Econometrics" American Economic Association 28 (28): 3-28, 2014

61 Susan Athey, "Beyond prediction: Using big data for policy problems" American Association for the Advancement of Science (AAAS) 355 (355): 483-485, 2017

62 Edward L. Glaeser, "BIG DATA AND BIG CITIES: THE PROMISES AND LIMITATIONS OF IMPROVED MEASURES OF URBAN LIFE" Wiley 56 (56): 114-137, 2018

63 Feigenbaum, James J., "Automated Census Record Linking"

64 Ford, Martin, "Architects of Intelligence: The Truth about AI from the People Building It" Packt Publishing 2018

65 Schapire, Robert E., "Adaptive Computation and Machine Learning Series" The MIT Press 2014

66 Schlkopf, Bernhard, "Adaptive Computation and Machine Learning Series" The MIT Press 2001

67 Lee, Kai-Fu, "AI Superpowers: China, Silicon Valley, and the New World Order" Houghton Mifflin Harcourt 2018